Implementing drug-gene rules

Pharmacogenomics is providing vital information about the interactions between medications and various genotypes. For clinicians, the challenge is accessing that complex information quickly in a clinical setting to prevent patients from experiencing possible adverse effects from medication.

Mayo Clinic is a pioneer in the implementation of drug-gene rules into clinical decision support systems. The Mayo-wide effort incorporates those rules in patients' electronic medical records (EMRs) and alerts clinicians to potential problems during patient visits.

"We need to provide physicians with the support and the systems that can guide decision-making in drug-gene interaction," says John Logan Black, M.D., a consultant in the Department of Laboratory Medicine and Pathology at Mayo Clinic in Rochester, Minnesota, and co-director of the Personalized Genomics Laboratory. "The interaction of even one gene with a medication is complex. As we expand to characterizing the effects of multiple genes, we will require sophisticated tools to help physicians."

Dr. Black spoke during a presentation on "Implementation of Drug-Gene Rules" at the 2014 Individualizing Medicine Conference at Mayo Clinic's campus in Minnesota. The conference was organized by the Mayo Clinic Center for Individualized Medicine.

EMR pop-ups

In 2012, Mayo Clinic formed a pharmacogenomics task force, then chaired by Dr. Black, to identify genetic variants with implications for drug therapy and to translate that information to clinical decision support systems. The task force interacts with the Mayo Pharmaceutical Formulary Committee and13 disease-oriented Mayo Clinic task forces as well as the Center for Individualized Medicine.

A key innovation from the task force is EMR messaging. If a patient is at risk of a drug-gene interaction, a warning appears when the clinician enters the prescribed medication into the patient's EMR. For example, a prescription for carbamazepine might generate an alert that the patient is positive for HLA-B*1502, or advice about testing for that genetic variant before beginning therapy. The concise warning is followed by a short explanatory paragraph, and a link to AskMayoExpert, an online clinical care tool with care process models, frequently asked questions and other information.

"The EMR messaging is very rich," Dr. Black says. "It can provide clinicians with information quickly."

Fitting the pieces together

Mayo Clinic has a representative on the Clinical Pharmacogenetics Implementation Consortium (CPIC), which addresses some of the barriers to implementation of pharmacogenetic tests in clinical practice. Speakers at the drug-gene rules presentation, which included other CPIC members, agreed that a viable pharmacogenetic clinical decision support system should be:

  • Timely — sending clinicians messages at the right time
  • Targeted — sent to clinicians, pharmacists and other allied health workers involved in care
  • Actionable — stating what drug should be avoided or given, and if given in what dosage
  • Clinician friendly — conveying information concisely to busy practitioners who often don't have substantial genetics training

"Implementing drug-gene rules into clinical decision support is like putting a puzzle together. Every piece has to fit correctly to have an efficient and effective rule for care providers," says Kelly K. Wix, Pharm.D., R.Ph., a medication management informaticist at Mayo Clinic's campus in Minnesota.

The panelists noted that pharmacogenetic testing is already being integrated into routine patient care. However, as the volume of information increases, clinical decision support systems become more important.

"As we move from genetic sequencing to offering whole-exome sequencing and whole-genome sequencing, there is going to be even more data," Dr. Black says. "Incorporating that data into the medical record gets truly complicated. But we are beginning to understand how to do that."